der8auer achieves a world record 5.8GHz overclock on AMD's Ryzen 1800X

[WYP]

The overclocker der8auer has achieved a 5.8GHz overclock on AMD's new Ryzen 7 1800X using ASUS' CrossHair VI Hero motherboard under LN2.



Read more on der8auer's world record Ryzen CPU overclock.

 

[Dawelio]

Haha nice! ... Although, I bet they used like liquid nitrogen or whatever it's called. Not regular AIO or similar

 

[Darkdayzzz]

Haha nice! ... Although, I bet they used like liquid nitrogen or whatever it's called. Not regular AIO or similar

Right at the top - The overclocker der8auer has achieved a 5.8GHz overclock on AMD's new Ryzen 7 1800X using ASUS' CrossHair VI Hero motherboard using LN2 (Liquid Nitrogen) cooling.

 

[Dawelio]

Right at the top - The overclocker der8auer has achieved a 5.8GHz overclock on AMD's new Ryzen 7 1800X using ASUS' CrossHair VI Hero motherboard using LN2 (Liquid Nitrogen) cooling.

Haha yes, do keep in mind that I wrote that post when only the headlines were in the post. The article nor the picture hadn't been linked yet.

 

[Bartacus]

He also delidded that chip. It appears delidding is essential to high overclocks with Ryzen. TTL even kinda proved this with his review. Those temps were pretty toasty. If a proper 240mm AIO can't cool this thing, a custom loop won't fare any better. Delidding also looks to be a scarier proposition with this thing due to how it's layed out. Still damn cool to see people push it! Even if the single core "gaming" performance isn't quite there yet, Ryzen is a hit in my books, and should ignite Intel into some serious competition, which bodes well for all of us.

 

[orbitalwalsh]

He also delidded that chip. It appears delidding is essential to high overclocks with Ryzen. TTL even kinda proved this with his review. Those temps were pretty toasty. If a proper 240mm AIO can't cool this thing, a custom loop won't fare any better. Delidding also looks to be a scarier proposition with this thing due to how it's layed out. Still damn cool to see people push it! Even if the single core "gaming" performance isn't quite there yet, Ryzen is a hit in my books, and should ignite Intel into some serious competition, which bodes well for all of us.

He delidded to show that there was no need for ryzen, its solder. Pitting Liquid metal would give worse result then leaving it as it comes

 

[AlienALX]

That's swick man

 

[Shingara]

How is it possible that an 1800x under ln2 can get the new world record but when tom did his, it was saying 80 degrees at 4.0 and his cooler was still not hot ?

 

[AlienALX]

How is it possible that an 1800x under ln2 can get the new world record but when tom did his, it was saying 80 degrees at 4.0 and his cooler was still not hot ?

Because they have delidded it. Then put it under LN2. So they have basically cut through a layer of alu and a layer of tin to get to the CPU core.

 

[Shingara]

Because they have delidded it. Then put it under LN2. So they have basically cut through a layer of alu and a layer of tin to get to the CPU core.

Thats not what i mean, unless the lids on ryzen are asbestos then the cooler should get very hot and be pumping out alot of heat not be luke warm when the cpu is sayings its 80+

 

[AlienALX]

Thats not what i mean, unless the lids on ryzen are asbestos then the cooler should get very hot and be pumping out alot of heat not be luke warm when the cpu is sayings its 80+

They are probably alu. They may be some sort of coated copper though. A lot of transfer plates are made from copper then coated. Back in the day when the guys used to lap CPUs they were copper.

As I say though, even a layer of alu is still another layer that heat needs to transfer through. It will also stay hot for longer for example than not having it there at all.

A couple of weeks ago a Ryzen hit 5.2, so it seems the delid did help.

 

[Shingara]

But isnt it a fact that when on ln2 it kills the temp sensor, thats why some boards have the hole in the back under the cpu for a temp probe. But lets say that the lids on ryzen arnt allowing heat through properly that the overclocks we get on water etc would be a translation of how far you could push it on ln2 aswell.

 

[AlienALX]

But isnt it a fact that when on ln2 it kills the temp sensor, thats why some boards have the hole in the back under the cpu for a temp probe. But lets say the the lids on ryzen arnt allowing heat through properly that it overclocks we get on water etc would be a translation of how far you could push it on ln2 aswell.

Any CPU will run better with no lid. The Athlon Tbirds had no lid either. So yeah, it's likely they would clock better with no lid at all on water.

However, they are very easy to damage. Loads of Athlons died because the sponge balancers on the corners came off and they basically fractured the die. That's why IHS were put on pretty much. Also, if your cooling fails with an IHS the CPU is usually fine. If it fails on a delid? watch this.

https://www.youtube.com/watch?v=kWTHHSE-Ej8

 

[Shingara]

well one less layer of anything is less thermal resistance, not that im gonna delid my 1800x, ever, its just this and other things are making me think there is something majorly wrong with how temps are being read at the mo, something just doesnt click.

And its the fact these got to that, obviously not disputing we arnt ever getting to those speeds on water lol, but the speeds we are getting now and the temps that are being posted is just iffy when its obvious ryzen has some chops to it.

 

[AlienALX]

well one less layer of anything is less thermal resistance, not that im gonna delid my 1800x, ever, its just this and other things are making me think there is something majorly wrong with how temps are being read at the mo, something just doesnt click.

And its the fact these got to that, obviously not disputing we arnt ever getting to those speeds on water lol, but the speeds we are getting now and the temps that are being posted is just iffy when its obvious ryzen has some chops to it.

It's an 8 core CPU dude. They get hot, fast.

It's also a very early gen 8 core CPU. Give AMD time? they will BIN down the temps etc and get higher clocking chips out there.

The original 5820k was a right dog. Every one went nuts for it because it was a "cheap" 6c 12t but most of the early samples barely scraped 4ghz. They would do 4.2ghz in furnace mode, but most of the guys who bought the early ones settled on 4ghz day to day.

Mine however? easy 4.5, temps not the problem. At 4.6 I can bench etc, but even with any amount of voltage it won't go any higher. Windows just does that sad face thing on the way in lol.

 

[Shingara]

It's an 8 core CPU dude. They get hot, fast.

It's also a very early gen 8 core CPU. Give AMD time? they will BIN down the temps etc and get higher clocking chips out there.

The original 5820k was a right dog. Every one went nuts for it because it was a "cheap" 6c 12t but most of the early samples barely scraped 4ghz. They would do 4.2ghz in furnace mode, but most of the guys who bought the early ones settled on 4ghz day to day.

Mine however? easy 4.5, temps not the problem. At 4.6 I can bench etc, but even with any amount of voltage it won't go any higher. Windows just does that sad face thing on the way in lol.

Im not disputing they get hot, i expect them to get hot, thats why you watercool, but the elephant in the room is that when they get hot that radiators push that heat out, if the rads arnt pushing heat out then either the heat isnt being xfered into the water cooling system/ the pump is dead or they arnt that hot at all.

 

[ketil92]

That voltage can't be right?

 

[AlienALX]

Im not disputing they get hot, i expect them to get hot, thats why you watercool, but the elephant in the room is that when they get hot that radiators push that heat out, if the rads arnt pushing heat out then either the heat isnt being xfered into the water cooling system or they arnt that hot at all.

It depends. If they are heat limited water cooling will help. If they are at their limits (like Fury X) then no amount of cooling will help.

And I strongly suspect it is the latter. When AMD say they broke their target and got 52% I would hazard a guess and say they got the extra 7% or whatever it was by pushing up the clocks.

With my 5820k I can not access Windows at 4.7ghz no matter how much voltage I shove through it.

 

[Shingara]

It depends. If they are heat limited water cooling will help. If they are at their limits (like Fury X) then no amount of cooling will help.

And I strongly suspect it is the latter. When AMD say they broke their target and got 52% I would hazard a guess and say they got the extra 7% or whatever it was by pushing up the clocks.

With my 5820k I can not access Windows at 4.7ghz no matter how much voltage I shove through it.

im not saying your wrong but my money is on they just arnt that hot and that the bios right now hasnt a scooby what todo with all those temp sensors and that its no where near the temps being reported. Even go as far to say i think when its reporting 80 thats its more likly only something like 50-60.

Because when lisa was on about xfr and how it was gonna be 4+ i was expecting 4.3-4.4 on an 1800x. (edit) on a really good custom water cooling loop.

 

[FluxTape]

He delidded to show that there was no need for ryzen, its solder. Pitting Liquid metal would give worse result then leaving it as it comes

Not quite correct. He said there might be gains because the solder layer is pretty thick.